​Hi Damjan,

I have tested the patch using ConnectX-4 EN (2x50G) with MLNOX_OFED 4.1.x on 
Xenial 16.04.2 (Kernel generic 4.11.12) with stable/17.07 branch

​Reason for kernel 4.11 was Mellanox requested 4.8 or later and other 
requirements pointed to 4.11, hence collapsing to the newer one.


Status:

a) With the patch I now stopped seeing the "UnknownInterface" but 50GE 
interface is seen as 100GE even when cabling a QSPF and kernel sees 40G speed.

localadmin@sm981:~$ sudo ethtool ens1f0
Settings for ens1f0:
Supported ports: [ FIBRE Backplane ]
Supported link modes:   1000baseKX/Full
                       10000baseKR/Full
                       40000baseKR4/Full
                       40000baseCR4/Full
                       40000baseSR4/Full
                       40000baseLR4/Full
Supported pause frame use: Symmetric Receive-only
Supports auto-negotiation: Yes
Advertised link modes:  1000baseKX/Full
                       10000baseKR/Full
                       40000baseKR4/Full
                       40000baseCR4/Full
                       40000baseSR4/Full
                       40000baseLR4/Full
Advertised pause frame use: No
Advertised auto-negotiation: Yes
Speed: 40000Mb/s
Duplex: Full
Port: FIBRE
PHYAD: 0
Transceiver: internal
Auto-negotiation: on
Supports Wake-on: d
Wake-on: d
Current message level: 0x00000004 (4)
      link
Link detected: yes


localadmin@sm981:~$ sudo vppctl show hardware-interfaces brief
              Name                Idx   Link  Hardware
HundredGigabitEthernet81/0/0       1     up   HundredGigabitEthernet81/0/0
HundredGigabitEthernet81/0/1       2     up   HundredGigabitEthernet81/0/1
local0                             0    down  local0


b) I get the following error message at VPP start

Jul 28 10:36:57 sm981 vpp[2750]: dpdk_bind_devices_to_uio:704: Unsupported PCI 
device 0x15b3:0x1013 found at PCI address 0000:81:00.1
Jul 28 10:36:57 sm981 vpp[2750]: dpdk_bind_devices_to_uio:704: Unsupported PCI 
device 0x15b3:0x1013 found at PCI address 0000:81:00.0

c) with kernel 4.11 I see VPP service stop hanging

localadmin@sm981:~$ sudo service vpp status​
● vpp.service - vector packet processing engine
   Loaded: loaded (/lib/systemd/system/vpp.service; enabled; vendor preset: 
enabled)
   Active: deactivating (stop-sigkill) (Result: timeout) since Fri 2017-07-28 
10:54:47 EDT; 1min 31s ago
  Process: 2743 ExecStartPre=/sbin/modprobe uio_pci_generic (code=exited, 
status=0/SUCCESS)
  Process: 2740 ExecStartPre=/bin/rm -f /dev/shm/db /dev/shm/global_vm 
/dev/shm/vpe-api (code=exited, status=0/SUCCESS)
 Main PID: 2750 (vpp_main)
   CGroup: /system.slice/vpp.service
           ‣ 2750 [vpp_main]

Jul 28 10:51:47 sm981 /usr/bin/vpp[2750]: unknown input `neigh
Jul 28 10:54:47 sm981 vpp[2750]: /usr/bin/vpp[2750]: unix_signal_handler:121: 
received signal SIGCONT, PC 0x7fb9a44477e0
Jul 28 10:54:47 sm981 vpp[2750]: /usr/bin/vpp[2750]: received SIGTERM, 
exiting...
Jul 28 10:54:47 sm981 systemd[1]: Stopping vector packet processing engine...
Jul 28 10:54:47 sm981 vpp[2750]: DPDK physical memory layout:
Jul 28 10:54:47 sm981 vpp[2750]: Segment 0: phys:0x3f0e800000, len:2097152, 
virt:0x7fb8e0c00000, socket_id:0, hugepage_sz:2097152, nchannel:0, nrank:0
Jul 28 10:54:47 sm981 vpp[2750]: Segment 1: phys:0x3f0ec00000, len:266338304, 
virt:0x7fb8a0e00000, socket_id:0, hugepage_sz:2097152, nchannel:0, nrank:0
Jul 28 10:54:47 sm981 vpp[2750]: Segment 2: phys:0x7f07c00000, len:2097152, 
virt:0x7fb8a0600000, socket_id:1, hugepage_sz:2097152, nchannel:0, nrank:0
Jul 28 10:54:47 sm981 vpp[2750]: Segment 3: phys:0x7f08000000, len:266338304, 
virt:0x7fb860800000, socket_id:1, hugepage_sz:2097152, nchannel:0, nrank:0
Jul 28 10:56:17 sm981 systemd[1]: vpp.service: State 'stop-sigterm' timed out. 
Killing.


Thanks,​


Daniel Bernier | Bell Canada

________________________________
From: Damjan Marion (damarion) <damar...@cisco.com>
Sent: Tuesday, July 4, 2017 8:08 PM
To: Bernier, Daniel
Cc: vpp-dev@lists.fd.io
Subject: Re: [vpp-dev] 50GE interface support on VPP

Hi Daniel,

Can you try with this patch?

https://gerrit.fd.io/r/#/c/7418/

Regards,

Damjan

On 4 Jul 2017, at 22:14, Bernier, Daniel 
<daniel.bern...@bell.ca<mailto:daniel.bern...@bell.ca>> wrote:

Hi,

I have ConnectX-4 50GE interfaces running on VPP and for some reason, they 
appear as “Unknown” even when running as 40GE.

localadmin@sm981:~$ lspci | grep Mellanox
81:00.0 Ethernet controller: Mellanox Technologies MT27700 Family [ConnectX-4]
81:00.1 Ethernet controller: Mellanox Technologies MT27700 Family [ConnectX-4]

localadmin@sm981:~$ ethtool ens1f0
Settings for ens1f0:
                Supported ports: [ FIBRE Backplane ]
                Supported link modes:   1000baseKX/Full
                                        10000baseKR/Full
                                        40000baseKR4/Full
                                        40000baseCR4/Full
                                        40000baseSR4/Full
                                        40000baseLR4/Full
                Supported pause frame use: Symmetric Receive-only
                Supports auto-negotiation: Yes
                Advertised link modes:  1000baseKX/Full
                                        10000baseKR/Full
                                        40000baseKR4/Full
                                        40000baseCR4/Full
                                        40000baseSR4/Full
                                        40000baseLR4/Full
                Advertised pause frame use: No
                Advertised auto-negotiation: Yes
                Speed: 40000Mb/s
                Duplex: Full
                Port: Direct Attach Copper
                PHYAD: 0
                Transceiver: internal
                Auto-negotiation: on
Cannot get wake-on-lan settings: Operation not permitted
                Current message level: 0x00000004 (4)
                                                       link
                Link detected: yes

localadmin@sm981:~$ sudo vppctl show interface
              Name               Idx       State          Counter          Count
UnknownEthernet81/0/0             1         up       rx packets                
723257
                                                     rx bytes                
68599505
                                                     tx packets                 
39495
                                                     tx bytes                 
2093235
                                                     drops                     
723257
                                                     ip4                        
48504
UnknownEthernet81/0/1             2         up       rx packets                
723194
                                                     rx bytes                
68592678
                                                     tx packets                 
39495
                                                     tx bytes                 
2093235
                                                     drops                     
723194
                                                     ip4                        
48504
local0                            0        down


Any ideas where this could be fixed?

Thanks,
----
Daniel Bernier | Bell Canada

_______________________________________________
vpp-dev mailing list
vpp-dev@lists.fd.io<mailto:vpp-dev@lists.fd.io>
https://lists.fd.io/mailman/listinfo/vpp-dev

_______________________________________________
vpp-dev mailing list
vpp-dev@lists.fd.io
https://lists.fd.io/mailman/listinfo/vpp-dev

Reply via email to